From: ao950 AT FreeNet DOT Carleton DOT CA (Paul Derbyshire) Newsgroups: comp.os.msdos.djgpp Subject: Re: SEVERE BUGS in ALLEGRO! Date: 28 Feb 1997 04:11:33 GMT Organization: The National Capital FreeNet Lines: 95 Message-ID: <5f5ltl$d3u@freenet-news.carleton.ca> References: <5etiq7$ir1 AT bertrand DOT ccs DOT carleton DOT ca> Reply-To: ao950 AT FreeNet DOT Carleton DOT CA (Paul Derbyshire) NNTP-Posting-Host: freenet5.carleton.ca To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp Shawn Hargreaves (Shawn AT talula DOT demon DOT co DOT uk) writes: > Paul Derbyshire writes: >>I have today discovered two very SEVERE BUGS in ALLEGRO that manifest >>themselves when the SOUND features are used. > > Wow. I have to say, your post leaves an ugly taste in my mouth. Not > because you are criticising my code (bug reports from users are > absolutely essential, and I'm normally delighted to be informed of any > problems that people are having), but because of your unpleasant and > confrontational attitude. I came very close to consigning your message > straight to the trash can (I don't like flame wars, so I have a policy > of not responding to any posts that make me feel personal animosity > towards the writer), but in the interests of getting these troubles > fixed, I'm willing to investigate your problems a bit further... Sorry, I was just a tad ticked that bugs this huge made it to a non-beta. (If the non-beta's like that, mayhap I should get 2.2beta...:)) >>SEVERE BUG #1: After running an Allegro program that executes a >>play_sample at some point, then returning to DOS, if you then run another >>DOS program that uses sound, e.g. the test-sbc program that comes with a >>creative labs soundblaster, the result is a squawk from the soundcard and >>a system hang. > > This is very probably a bug. I need a bit more info from you, though, in > particular: > > What soundcard do you have? Model, irq number, dma channel, etc... Model: Regular Sound Blaster. IRQ 2. DMA 1. > What is your BLASTER environment variable set to? SET BLASTER = A220 I2 D1 > Which sound driver is Allegro using? (displayed by play.exe) Sound blaster... > Does the problem happen if you only use the MIDI sound, or only the > digital sound, or can either cause it? It was play_sample. Since I haven't managed to find a single damn piece of software on the whole wide Internet to compose midi music, I haven't had opportunity to test the midi functions yet. > Does the problem occur when you re-run the Allegro program, or only when > you run other (eg. creative labs) code? Other code. If I run an Allegro program it's OK. If I run Windows, though it uses the sound, that's OK. Creative labs code...kaboom. :P It also happens only intermittently, and I vaguely think it might be only if the program is aborted unusually, eg by a gpf or such, before exit() is called or main falls through. > Are you running under win95, or clean DOS? Essentially clean DOS: Win 95 with an F8 menu boot "command prompt only", specifically because the "MS-DOS Mode" seems to have traces of Windows in the system. LFN support is turned off. >>SEVERE BUG #2: load_sample always returns NULL, even if the string it is >>sent as parameter is the fille name of a mono WAV exactly as per > > What version of Allegro is this? Up until version 2.2, there were some > problems with the WAV parser (it only understood a small subset of the > RIFF format, and didn't like the samples produced by win95 utilities). > If you have an earlier Allegro version, get 2.2 from my web page and see > if that clears the trouble up. If you already have 2.2, please send me > one of the WAV's that you can't read, and I'll investigate further... I shall check out 2.2...thanks. >>I cannot imagine that either of these two bugs could have gone completely >>undiscovered. There is no way you didn't know about these when you >>released the version of allegro I'm using. Yet the documentation makes no > > I can't remember who it was that said "never attribute to malice what > can be adequately explained by stupidity", but it's a very true > statement :-) Did it never occur to you that I might actually have > tested this code, and that the reason I didn't notice these problems was > that they don't occur on my machine? Ah. of course, any developer has only a subset of all possible hardware and software configurations for testing. And your WAV parser must have been tested only on a sampling of wav files that didn't happen to include all the possibilities...hm. -- .*. Where feelings are concerned, answers are rarely simple [GeneDeWeese] -() < When I go to the theater, I always go straight to the "bag and mix" `*' bulk candy section...because variety is the spice of life... [me] Paul Derbyshire ao950 AT freenet DOT carleton DOT ca, http://chat.carleton.ca/~pderbysh